[gmx-users] Multiple runs for same system

[Justin Lemkul]

On 2/25/18 6:07 PM, Dr. Seema Mishra wrote:
> Thanks Justin. Meaning I will simply set gen_seed as -1 instead of 173529. Right?

However you want to do it, either by specifying a seed or letting grompp 
generate one. Some prefer the former for reproducibility, but given all 
the things that can affect binary reproducibility, I don't think it's a 
big concern.

> After this, how many runs will be performed? One? Then after the production run is finished, start it again with gen_seed=-1 for the second run?

Each .tpr specifies one initial set of conditions for a simulation. If 
you want another simulation, generate another .tpr that specifies a 
different set of conditions.

-Justin

> This is to be done for 3 independent runs for same protein-ligand system.
>
> Seema
>   
>
>      On Monday, 26 February 2018 4:16 AM, Justin Lemkul <jalemkul at vt.edu> wrote:
>   
>
>   
> Multiple runs are typically initiated simply by changing gen_seed when
> generating velocities at the outset of equilibration.
>
> As for clustering, you'll have to explain what you mean. Clustering
> ligand poses? Protein? What? Does a Google search for "GROMACS
> clustering" or the help information for gmx cluster point you to
> anything useful
>
> -Justin
>
> On 2/25/18 4:11 PM, Dr. Seema Mishra wrote:
>> Or just direct me to the steps written anywhere....
>>
>>        On Saturday, 24 February 2018 3:51 PM, Dr. Seema Mishra <seema_uoh at yahoo.com> wrote:
>>    
>>
>>    Hi, Can anyone tell me the steps and commands for performing multiple runs of 50 ns for same protein-ligand system? Also the clustering for further analyses. Thanks.
>>
>>
>>      

-- 
==================================================

Justin A. Lemkul, Ph.D.
Assistant Professor
Virginia Tech Department of Biochemistry

303 Engel Hall
340 West Campus Dr.
Blacksburg, VA 24061

jalemkul at vt.edu | (540) 231-3129
http://www.biochem.vt.edu/people/faculty/JustinLemkul.html

==================================================